Site of cochlear stimulation and its effect on electrically evoked compound action potentials using the MED-EL standard electrode array
© Brill et al; licensee BioMed Central Ltd. 2009
Received: 26 May 2009
Accepted: 16 December 2009
Published: 16 December 2009
The standard electrode array for the MED-EL MAESTRO cochlear implant system is 31 mm in length which allows an insertion angle of approximately 720°. When fully inserted, this long electrode array is capable of stimulating the most apical region of the cochlea. No investigation has explored Electrically Evoked Compound Action Potential (ECAP) recordings in this region with a large number of subjects using a commercially available cochlear implant system. The aim of this study is to determine if certain properties of ECAP recordings vary, depending on the stimulation site in the cochlea.
Recordings of auditory nerve responses were conducted in 67 subjects to demonstrate the feasibility of ECAP recordings using the Auditory Nerve Response Telemetry (ART™) feature of the MED-EL MAESTRO system software. These recordings were then analyzed based on the site of cochlear stimulation defined as basal, middle and apical to determine if the amplitude, threshold and slope of the amplitude growth function and the refractory time differs depending on the region of stimulation.
Findings show significant differences in the ECAP recordings depending on the stimulation site. Comparing the apical with the basal region, on average higher amplitudes, lower thresholds and steeper slopes of the amplitude growth function have been observed. The refractory time shows an overall dependence on cochlear region; however post-hoc tests showed no significant effect between individual regions.
Obtaining ECAP recordings is also possible in the most apical region of the cochlea. However, differences can be observed depending on the region of the cochlea stimulated. Specifically, significant higher ECAP amplitude, lower thresholds and steeper amplitude growth function slopes have been observed in the apical region. These differences could be explained by the location of the stimulating electrode with respect to the neural tissue in the cochlea, a higher density, or an increased neural survival rate of neural tissue in the apex.
The Clinical Investigation has the Competent Authority registration number DE/CA126/AP4/3332/18/05.
Objective measures are a widely used and valuable tool in the field of cochlear implants (CIs). During surgery they provide first indications of successful implantation and after surgery they are used to facilitate the individual fitting of stimulation parameters. One objective measure of the auditory nerve's response to stimulation is the Electrically Evoked Compound Action Potential (ECAP). This response is particularly advantageous because it allows the clinician to directly measure auditory nerve fibre potentials on implanted patients. In analyzing the physiological response to the electrical stimulation transmitted by the implant, information can be obtained regarding the expected and actual function of the peripheral nerve. This information can be used intraoperatively to adjust the placement of the intracochlear electrode and for technical functional testing. Postoperatively, the ECAP recordings can be used to measure the neuronal potentials elicited by electrode stimulation along the basilar membrane. In addition, these measurements may help to determine the upper and lower limits of the stimulation current [the hearing threshold (THR) and the maximum comfort level (MCL)] in cases of fitting [1–7] or to evaluate the stimulation current field along the cochlea and the interaction of individual electrodes [8–10].
Stimulation of an intracochlear electrode results in the excitation of specific populations of neural fibres that are distributed along the basilar membrane. Previous investigations have been limited in terms of how much of the cochlea can be analyzed because of maximal insertion depths and insertion angles of the electrode arrays. For this investigation, the Auditory Nerve Telemetry (ART)  in the MED-EL MAESTRO CI system (MED-EL, Innsbruck, Austria) was used to perform ECAP recordings. The MED-EL standard electrode array allows an insertion depth of 31 mm or 720° and comprises 12 channels , allowing access to the most apical region of the cochlea. The channels are spaced 2.4 mm apart, and are numbered E 1 to E 12, from apical to basal.
The primary objective of this study was to establish whether a difference exists in ECAP properties depending on the region of the cochlea that is stimulated and measured. Previous studies using other CI systems found increased ECAP amplitudes at the apical electrodes [13, 14]. Note that these publications did not realize a full cochlear coverage. Also electrophysiological models did not analyze a (deep) apical stimulation, comparing it with a middle or basal stimulation. Therefore in this study, an analysis of amplitude, threshold, and slope of the ECAP recordings obtained in various regions of cochlear stimulation was conducted. A secondary objective of this study was to determine if the ECAP amplitude in relative refractory state shows systematic differences for different stimulation and recording sites.
experienced and 21 inexperienced CI users were recruited from 13 centres in Germany for a clinical study [15, 16] investigating CI coding strategies and ECAP recordings. The study was approved by the International Freiburger Ethic Commission (FECI 05/2134). Of these 67 CI subjects, 34 were women and 33 were men. The mean age at onset of hearing loss was 43.4 years with a range of 1 to 73 years. The mean age at implantation was 55.4 years with a range of 20 to 76 years. 37 subjects were implanted on the left side and 30 were implanted on the right. All subjects were postlingually deaf, (defined as an onset of severe-to-profound sensorineural hearing loss occurring after 6 years of age) and were unilaterally implanted with a MED-EL PULSARci100 or SONATAti100 using the standard electrode array. All subjects were required to have at least 10 active electrodes at the last fitting. All subjects were native German speakers.
Experienced subjects were adults (18 years or older) with a minimum of six months of device experience (mean = 1.4 years; range = 7 to 31 months). The inexperienced subjects were adults who received their first CI after having undergone a hearing aid trial where minimal aided benefit from hearing aid(s) was determined. These subjects fell within the medical and audiological guidelines established by their respective centres for cochlear implantation. Apart from experience with the device, the inclusion criteria were the same for both the experienced and inexperienced groups.
If for one subject more than one recording of an ECAP property was available, the mean value was used in the analysis. To test the hypotheses, a general linear model was applied and analyses of variance (ANOVA) for repeated measurements  with the region as factor were performed for each test condition. If the assumption of sphericity was not tenable according to Mauchly's test , the Greenhouse-Geisser correction  was applied. To detect significant effects of the region on the ECAP measurement, parametric paired Student's t-tests were used. After adjustments for multiple comparisons with the use of Bonferroni's procedure  p-values of less than 0.017 were considered to indicate statistical significance. The software employed for the statistical analysis was Matlab Rev 188.8.131.5220 with the Statistics Toolbox Version 5.0.
ECAP measurements were performed as part of a study [15, 16] evaluating ECAP recordings, and the performance of subjects with the Fine Structure Processing (FSP) strategy as improved in the OPUS audio processors, acutely and after three months of device use. ECAPs were recorded postoperatively using the MAESTRO 2.0 system software connected to a DIB II interface box. For the experienced subjects ECAP recordings were obtained at acute switch-over from their clinical TEMPO+ speech processor to the OPUS 1 speech processor employing the FSP coding strategy, and for the inexperienced subjects at initial stimulation. If no ECAP could be recorded from a subject of either group, measurements were reattempted three months after the initial test date.
For all measurements, the stimulation artifact was removed using an alternating stimulation approach. Each measurement is performed twice, with a cathodic/anodic and an anodic/cathodic stimulation pulse, respectively. When averaging the two measurements, the stimulation artifact vanishes and the ECAP signal remains. The recording artifact was removed subtracting a zero amplitude template .
Mapping of the electrodes to the cochlear region depending on the electrode insertion depth.
Active Electrodes: 1-12
Active Electrodes: 1-11
Active Electrodes: 1-10
The ECAP measurement results reported as mean values with the standard deviation.
220.8 ± 114.4 μV
257.9 ± 129.7 μV
341.3 ± 200.5 μV
273.3 ± 159.6 μV
356.2 ± 114.0 cu
337.8 ± 100.2 cu
307.3 ± 113.7 cu
333.8 ± 110.3 cu
ECAP amplitude growth function
0.747 ± 0.389 μV/cu
0.766 ± 0.384 μV/cu
1.092 ± 0.638 μV/cu
0.869 ± 0.506 μV/cu
ECAP amplitude recovery sequence
163.3 ± 87.7 μV
192.8 ± 70.2 μV
223.4 ± 104.2 μV
193.2 ± 90.1 μV
1267.9 ± 301.9 μs
1236.0 ± 415.6 μs
1519.0 ± 431.8 μs
1341.0 ± 400.1 μs
Observed regional effects on the respective ECAP features.
P < 0.001
F = 10.994
p = 0.0007
p = 0.0035
p = 0.0527
ε = 0.748
P = 0.003
F = 5.376
p = 0.0026
p = 0.0511
p = 0.2314
ε = 0.999
ECAP slope (ampl. growth function)
P < 0.001
F = 12.038
p = 0.0014
p = 0.0002
p = 0.7055
ε = 0.690
ECAP amplitude (recovery function)
P = 0.045
F = 3.460
p = 0.0086
p = 0.2770
p = 0.1775
Interpulse nterval (rIPI)
P = 0.029
F = 4.015
p = 0.0590
p = 0.0262
p = 0.7198
Ecap Amplitude Growth Sequence
ECAP measurements were possible for 58 users accounting for 86.6% of all subjects. Within our subjects, the presence or absence of ECAP recordings varied between ECAP recordings sites. The presence of a clear response was greatest when the stimulating electrode and measuring electrode were in the middle and apical region of the cochlea. Recordings from the middle region were obtained in 52 (77.6%) subjects, and from the apical region in 51 (76.1%). In the basal region, responses were obtained in 38 subjects (56.7%). The small number of subjects who had responses in the basal region reduced the number of subjects for whom ECAPs were successfully measured in all three regions to 34 users (50.7%). Data reporting is therefore based on those subjects (N = 34) except where otherwise stated.
Ecap Recovery Sequence
Recovery sequences were measured in the experienced group (N = 46). Unlike for the amplitude growth sequence, variability in the presence or absence of a clear response depending on the cochlear region was not seen. Clear responses were recorded for 21 subjects (45.7%) in the basal region, for 25 subjects (54.3%) in the middle region, and for 21 subjects (45.7%) in the apical region. There were 16 subjects (34.8%) for whom an ECAP recovery sequence could be measured in all three regions and 29 subjects (63.0%) for whom an ECAP recovery sequence determination was possible in any region.
Ecap Amplitude Using Amplitude Growth Sequence
Ecap Threshold Using Amplitude Growth Sequence
Slope of Ecap Amplitude Growth Function
Ecap Amplitude Derived Using Recovery Sequence
Recovery Inter-Pulse Interval (rIPI)
In this study we examined if certain properties of ECAP recordings vary, depending on the stimulation site in the cochlea. To various degrees, we found a significant effect of stimulation site on ECAP amplitude, ECAP threshold, slope of ECAP amplitude growth function and ECAP recovery inter-pulse interval (Table 2, Table 3).
The significant increase in ECAP amplitude towards the apical region could be explained by a narrowed distance between the recording electrode and the stimulated neural tissue (because of reduced diameter of the cochlear turns towards the apex). Another possible factor is the neural tissue itself. A greater density or survival rate of neuronal tissue adjacent to the electrode in the apical region could also explain the increase in ECAP amplitude which is also supported by the steeper growth function observable in Figure 5. Better neural survival of neural structures in the apex might be produced by later deafening of the apex, e.g. in many cases of progredient deafness. Differences in impedances of the electrode-tissue interfaces between cochlear regions can not explain this effect because the implant uses current sources and high-impedance recording circuitry and is therefore not affected by different electrode-tissue interface impedances.
We also investigated for the analyzed subjects whether an increased stimulation amplitude towards the apical region might be responsible for the amplitude increase. For the basal region the mean stimulation amplitude was 654.0 ± 224.4 μV. For the middle and apical region, mean stimulation amplitudes were 723.1 ± 293.8 μV and 706.9 ± 317.1 μV, respectively. An ANOVA showed that the stimulus amplitudes are not dependent of the measurement region (p = 0.244, F = 1.445). As stimulation amplitudes at initial stimulation usually differ from amplitudes for experienced users, we did also an ANOVA for the experienced users only, which in contrast to the data above showed a region dependency of the stimulation amplitudes (p = 0.001, F = 7.730). Paired t-tests show that there is a significant difference between the basal and middle (p = 0.001) as well as the basal and apical region (p = 0.004). This could partially have contributed to the ECAP amplitude increase towards the apex. The middle and apical region, however, shows no significant difference (p = 0.812) in stimulation amplitude although a significant difference in ECAP amplitude was found here also (p = 0.004), as mentioned above. Therefore the chosen stimulation amplitude can only partly explain the region dependency of the ECAP amplitude. This is confirmed by the fact that studies using other CI systems also found increased ECAP amplitudes for more apical electrodes [13, 14].
The significant decrease in ECAP threshold towards the apex could again be attributed to the narrowed distance between the recording electrode and the surrounding tissue. Additionally, similar to above, a greater density of neuronal tissue or survival rate adjacent to the electrode could also lead to a reduction in the stimulation amplitude required to trigger an action potential. As mentioned previously, regional differences in the impedance of the electrodes cannot explain the decrease towards the apex.
The slope of the ECAP growth function should strongly correlate with an increase in the number of neurons that respond to every increment in stimulation level. The steeper growth function shown in Figure 5 indicates that towards the apical region, a greater number of neural elements are activated for every increment in stimulation level. Since spiral ganglion cells do not extend into the apical region of the cochlea, these neural elements should mainly be afferent peripheral axons.
As the ECAP amplitudes from the recovery sequences did not differ significantly from those from growth sequences, ECAP amplitudes from recovery sequences also showed a region dependency effect. The significant difference in ECAP amplitudes between the basal and apical region is also found here. The fact that - in contrast to ECAP amplitudes from growth sequences - no significant difference between the middle and apical region could be found here is presumably due to the smaller subject group (16 instead of 34) in which recovery sequences were performed.
The recovery interpulse interval (rIPI) shows an overall effect but no significant post-hoc differences for different stimulation and recording sites. There are at least two different mechanisms that could lead to changes in the latency of ECAP potentials along the cochlea: Some firing features of type II spiral ganglion neurons are cochlear region dependent; the latency is reported to be longer for the apical region [22, 23] (determined in murine cochlea). Secondly, the latency depends on the site of stimulation and recording sites. These locations depend on the spiral ganglion cell arrangement between apex and base that is reported to be different [24, 25] (human cochlea). This effect is also assumed to be the reason for the double P peaks, seen in 9.5% of ECAP responses .
The observed increase in ECAP amplitude towards the apex of the cochlea adds another aspect to the discussion about complete electrode insertion in cochlear implantation. We interpret this result as further evidence for the usefulness of apical cochlear stimulation. The data from objective measurements presented here complement data from behavioural assessment which show that speech discrimination improves when the most apical region of the cochlear is stimulated along with the medial and basal region [27, 12]. Restricting the most apical stimulation to 300° to 400° insertion angle does not take advantage of this portion of the cochlea.
Four different properties of the ECAP in adult subjects were analysed based on the region of the cochlea where the response was generated. All four properties analysed were found to be affected by the cochlear region. Apical recordings showed on average higher ECAP amplitudes, lower ECAP thresholds, and steeper slopes of the ECAP amplitude growth function. Also the ECAP refractory time showed a significant effect of stimulation region. These regional differences could be due to the closer proximity of the stimulating electrode to neural tissue in the apex and/or to a higher density or survival rate of neural tissue in the apex, which could also explain the robustness of the apical response and the steeper growth function. The recovery inter-pulse interval showed an overall dependence on cochlear region while significant effects between the individual regions could not be shown.
The novel available large number of ECAP measurements from the most apical region of the cochlea show that significant differences exist between the apical and the basal region. The apical ECAP recordings show a clear response with higher (on average) amplitudes and a lower threshold both of which are advantageous in terms of measurement success. These findings imply that future studies being conducted on ECAPs should include, wherever possible, a regional analysis of the results to incorporate any region dependent effects.
The authors would like to thank the following persons who contributed to the study by providing study subjects or assisting in the process: Mr. Steinhoff from the Klinikum rechts der Isar of the TU München; Dr. Suckfull from the Klinikum Großhadern; Prof. Dr. Gstöttner from the Klinikum JW. Goethe-Universität; PD Dr. Mürbe from the Universitätsklinikum Carl Gustav Carus; Prof. Dr. Esser from the HELIOS Klinikum Erfurt GmbH; Dr. Wesarg from the Universitätsklinik Freiburg; Dr. Maier from the Universitätsklinikum Hamburg-Eppendorf; Dr. Büchner from the Medizinische Hochschule Hannover; Dr Schelhoru-Neide from the Universitätsklinikum Jena; Prof. Dr. von Specht from the Otto-von-Guericke-Universität Magdeburg; Prof. Dr. Pau from the Universtität Rostock; Prof. Dr. Zenner and Dr. Tropitisch from the Universitäts-HNO-Klinik Tübingen; Mr. Leiacker from the Universitätsklinikum Ulm. We further thank Melissa Waller for here comments on the paper, Philipp Spitzer who analyzed the recovery sequences, Martina Deibl who performed the initial statistical analysis and Robin Cooley who provided medical writing services on behalf of the MED-EL GmbH.
- Brown CJ, Abbas PJ, Gantz BJ: Preliminary experience with neural response telemetry in the nucleus CI24 M cochlear implant. Am J Otol 1998,19(3):320–327.Google Scholar
- Brown CJ, Hughes ML, Luk B, Abbas PJ, Wolaver A, Gervais J: The relationship between EAP and EABR thresholds and levels used to program the nucleus 24 speech processor: data from adults. Ear Hear 2000,21(2):151–163. 10.1097/00003446-200004000-00009View ArticleGoogle Scholar
- Franck KH: A model of a nucleus 24 cochlear implant fitting protocol based on the electrically evoked whole nerve action potential. Ear Hear 2002,23(1 Suppl):67S-71S. 10.1097/00003446-200202001-00008View ArticleGoogle Scholar
- Franck KH, Norton SJ: Estimation of psychophysical levels using the electrically evoked compound action potential measured with the neural response telemetry capabilities of Cochlear Corporation's CI24 M device. Ear Hear 2001,22(4):289–299. 10.1097/00003446-200108000-00004View ArticleGoogle Scholar
- Hughes ML, Vander Werff KR, Brown C, Abbas P, Kelsay D, Teagle H, Lowder MW: A longitudinal study of electrode impedance, the electrically evoked compound action potential, and behavioral measures in nucleus 24 cochlear implant users. Ear & Hearing 2001, 22: 471–486. 10.1097/00003446-200112000-00004View ArticleGoogle Scholar
- Hochmair-Desoyer I, Schulz E, Moser L, Schmidt M: The HSM sentence test as a tool for evaluating the speech understanding in noise of cochlear implant users. Am J Otol 1997,18(6 Suppl):S83.Google Scholar
- Zimmerling M, Hochmair ES: EAP recordings in ineraid patients - correlations with psychophysical measures and possible implications for patient fitting. Ear & Hearing 2002, 23: 81–91. 10.1097/00003446-200204000-00001View ArticleGoogle Scholar
- Abbas PJ, Hughes ML, Brown CJ, Miller CA: Channel interaction in cochlear implant users evaluated using the electrically evoked compound action potential. Audiology & Neuro-Otology 2004,9(4):203–213. 10.1159/000078390View ArticleGoogle Scholar
- Cohen LT, Richardson LM, Saunders E, Cowan RSC: Spatial spread of neural excitation in cochlear implant recipients: comparison of improved ECAP method and psychophysical forward masking. Hearing Research 2003, 179: 72–87. 10.1016/S0378-5955(03)00096-0View ArticleGoogle Scholar
- Cohen LT, Saunders E, Richardson LM: Spatial spread of neural excitation: comparison of compound action potential and forward-masking data in cochlear implant recipients. Int J Audiol 2004,43(6):346–355. 10.1080/14992020400050044View ArticleGoogle Scholar
- Zierhofer CM: Multichannel cochlear implant with neural response telemetry. US Patent 2003.Google Scholar
- Hochmair I, Arnold W, Nopp P, Jolly C, Müller J, Roland P: Deep electrode insertion in cochlear implants: apical morphology, electrodes and speech perception results. Acta Otolaryngol 2003,123(5):612–617. 10.1080/000164803100001844View ArticleGoogle Scholar
- Polak M, Hodges A, King J, Balkany T: Further prospective findings with compound action potentials from Nucleus 24 cochlear implants. Hearing Research 2004, 188: 104–116. 10.1016/S0378-5955(03)00309-5View ArticleGoogle Scholar
- Frijns JHM, Briaire JJ, de Laat JA, Grote JJ: Initial evaluation of the Clarion CII cochlear implant: speech perception and neural response imaging. Ear & Hearing 2002, 23: 184–197. 10.1097/00003446-200206000-00003View ArticleGoogle Scholar
- Anderson I, Deibl M: FS1 Clinical investigation CRD2005CIP001: experienced users final report. 2007.Google Scholar
- Anderson I, Deibl M: FS1 Clinical investigation CRD2005CIP001: inexperienced users final report. 2007.Google Scholar
- Marques de Sá J: Applied statistics: Using SPSS, STATISTICA, MATLAB and R. 2nd edition. Berlin: Springer; 2007.View ArticleGoogle Scholar
- Mauchly J: Significance test for sphericity of a normal n-variate distribution. Annals of Mathematical Statistics 1940, 11: 204–209. 10.1214/aoms/1177731915MathSciNetView ArticleGoogle Scholar
- Greenhouse S, Geisser S: On methods in the analysis of profile data. Psychometrika 1959,24(2):95–112. 10.1007/BF02289823MathSciNetView ArticleGoogle Scholar
- Bortz J: Statistik für Human- und Sozialwissenschaftler. Berlin: Springer; 2005.Google Scholar
- Morsnowski A, Charasse B, Collet L, Killian M, Müller-Deile J: Measuring the refractoriness of the electrically stimulated auditory nerve. Audiol Neurootol 2006,11(6):389–402. 10.1159/000095966View ArticleGoogle Scholar
- Adamson CL, Reid MA, Mo ZL, Bowne-English J, Davis RL: Firing features and potassium channel content of murine spiral ganglion neurons vary with cochlear location. J Comp Neurol 2002,447(4):331–350. 10.1002/cne.10244View ArticleGoogle Scholar
- Reid MA, Flores-Otero J, Davis RL: Firing patterns of type II spiral ganglion neurons in vitro. J Neurosci 2004,24(3):733–742. 10.1523/JNEUROSCI.3923-03.2004View ArticleGoogle Scholar
- Glueckert R, Pfaller K, Kinnefors A, Rask-Andersen H, Schrott-Fischer A: The Human Spiral Ganglion: New insights into ultrastructure, survival rate and implications for cochlear implants. Audiology & Neuro-Otology 2005, 10: 258–273. 10.1159/000086000View ArticleGoogle Scholar
- Glueckert R, Pfaller K, Kinnefors A, Schrott-Fischer A, Rask-Andersen H: High resolution scanning electron microscopy of the human organ of Corti. A study using freshly fixed surgical specimens. Hearing Research 2005,199(1–2):40–56.Google Scholar
- Lai WK, Dillier N: A Simple Two-Component Model of the Electrically Evoked Compound Action Potential in the Human Cochlea. Audiology & Neuro-Otology 2000, 5: 333–345. 10.1159/000013899View ArticleGoogle Scholar
- Hamzavi J, Arnoldner C: Effect of deep insertion of the cochlear implant electrode array on pitch estimation and speech perception. Acta Otolaryngol 2006,126(11):1182–1187. 10.1080/00016480600672683View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.